Abstract
Human platelet alloantigens (HPAs) are caused by single nucleotide polymorphisms within functionally important platelet membrane glycoproteins. Antibodies against HPAs have been implicated in several clinically important alloimmune platelet disorders, including neonatal alloimmune thrombocytopenia (NAIT), post-transfusion purpura, and platelet transfusion refractoriness. A single C29523T nucleotide substitution, resulting in a Leu33Pro amino acid polymorphism within PSI domain of the integrin β3 subunit is responsible for generating the HPA-1a/HPA-1b alloantigenic epitopes, and is the most commonly implicated HPA in alloimmune platelet disorders in Caucasian populations. Allogeneic mismatch between an HPA-1b homozygous mother and an HPA-1a-positive fetus can lead to maternal alloimmunization, with the production of anti-HPA-1a antibodies that cross the placenta, where they bind fetal platelets and endothelial cells. The resulting thrombocytopenia and damage to fetal endothelial cells results in NAIT. Existing animal models of NAIT involve either (1) injecting human platelets into NOD/SCID immunodeficient mice followed by intraperitoneal injections of human maternal alloantisera, or (2) transfusing wild-type mouse platelets into GPIIb-IIIa-deficient mice to generate an isoimmune, rather than an alloimmune, response. Neither of these models truly mimic NAIT, as only the kinetics of platelet clearance can be measured in the former model, while the latter is a model of iso-, rather than allo-, immunity. Mouse β3 contains neither Leu nor proline at residue 33 position, and human maternal anti-HPA-1a antibodies do not recognize the murine protein. Previous studies have shown, however, that substitution of only four amino acids within the PSI domain is sufficient for reconstituting the human anti-HPA-1a epitope into the murine protein. This laid the foundation for rational genomic editing to recreate the HPA-1a epitope in mice. Guide RNAs that target the ITGB3 gene near the region encoding amino acid 33 were cloned into the CRISPR expression plasmid, px459. A 200 bp single-stranded DNA oligonucleotide corresponding to murine β3, but containing nine nucleotide substitutions that introduced four amino acid substitutions (T30A, S32P, Q33L, and N39D) into the mouse β3 cDNA was used as a template for homology-directed repair (HDR) of the endogenous ITGB3 gene. These four amino acid substitutions (APLD) have previously been shown to be necessary for reconstituting the human anti-HPA-1a epitope in murine β3. The CRISPR plasmid and HDR repair template were co-injected into fertilized mouse zygotes, the progeny of which were then screened by diagnostic restriction enzyme digestion for successful homozygous integration of the HDR repair template. That platelets from the APLD mice expressed the human HPA-1 alloantigenic epitope was confirmed by flow-cytometry and Western-blot analysis using the HPA-1a-selective murine monoclonal antibody, SZ21, as well as human maternal anti-HPA-1a alloantisera. Successfully generation of mice whose platelets and endothelial cells express the human HPA-1a alloantigenic epitope will permit development of an authentic murine model of NAIT that should facilitate the investigation of a number of clinically-important issues in the diagnosis and treatment of platelet alloimmune disorders.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.